1 Open vSwitch <http://openvswitch.org>
3 Frequently Asked Questions
4 ==========================
9 Q: What is Open vSwitch?
11 A: Open vSwitch is a production quality open source software switch
12 designed to be used as a vswitch in virtualized server environments. A
13 vswitch forwards traffic between different VMs on the same physical host
14 and also forwards traffic between VMs and the physical network. Open
15 vSwitch supports standard management interfaces (e.g. sFlow, NetFlow,
16 RSPAN, CLI), and is open to programmatic extension and control using
17 OpenFlow and the OVSDB management protocol.
19 Open vSwitch as designed to be compatible with modern switching
20 chipsets. This means that it can be ported to existing high-fanout
21 switches allowing the same flexible control of the physical
22 infrastructure as the virtual infrastructure. It also means that
23 Open vSwitch will be able to take advantage of on-NIC switching
24 chipsets as their functionality matures.
26 Q: What virtualization platforms can use Open vSwitch?
28 A: Open vSwitch can currently run on any Linux-based virtualization
29 platform (kernel 2.6.18 and newer), including: KVM, VirtualBox, Xen,
30 Xen Cloud Platform, XenServer. As of Linux 3.3 it is part of the
31 mainline kernel. The bulk of the code is written in platform-
32 independent C and is easily ported to other environments. We welcome
33 inquires about integrating Open vSwitch with other virtualization
36 Q: How can I try Open vSwitch?
38 A: The Open vSwitch source code can be built on a Linux system. You can
39 build and experiment with Open vSwitch on any Linux machine.
40 Packages for various Linux distributions are available on many
41 platforms, including: Debian, Ubuntu, Fedora.
43 You may also download and run a virtualization platform that already
44 has Open vSwitch integrated. For example, download a recent ISO for
45 XenServer or Xen Cloud Platform. Be aware that the version
46 integrated with a particular platform may not be the most recent Open
49 Q: Does Open vSwitch only work on Linux?
51 A: No, Open vSwitch has been ported to a number of different operating
52 systems and hardware platforms. Most of the development work occurs
53 on Linux, but the code should be portable to any POSIX system. We've
54 seen Open vSwitch ported to a number of different platforms,
55 including FreeBSD, Windows, and even non-POSIX embedded systems.
57 By definition, the Open vSwitch Linux kernel module only works on
58 Linux and will provide the highest performance. However, a userspace
59 datapath is available that should be very portable.
61 Q: What's involved with porting Open vSwitch to a new platform or
64 A: The PORTING document describes how one would go about porting Open
65 vSwitch to a new operating system or hardware platform.
67 Q: Why would I use Open vSwitch instead of the Linux bridge?
69 A: Open vSwitch is specially designed to make it easier to manage VM
70 network configuration and monitor state spread across many physical
71 hosts in dynamic virtualized environments. Please see WHY-OVS for a
72 more detailed description of how Open vSwitch relates to the Linux
75 Q: How is Open vSwitch related to distributed virtual switches like the
76 VMware vNetwork distributed switch or the Cisco Nexus 1000V?
78 A: Distributed vswitch applications (e.g., VMware vNetwork distributed
79 switch, Cisco Nexus 1000V) provide a centralized way to configure and
80 monitor the network state of VMs that are spread across many physical
81 hosts. Open vSwitch is not a distributed vswitch itself, rather it
82 runs on each physical host and supports remote management in a way
83 that makes it easier for developers of virtualization/cloud
84 management platforms to offer distributed vswitch capabilities.
86 To aid in distribution, Open vSwitch provides two open protocols that
87 are specially designed for remote management in virtualized network
88 environments: OpenFlow, which exposes flow-based forwarding state,
89 and the OVSDB management protocol, which exposes switch port state.
90 In addition to the switch implementation itself, Open vSwitch
91 includes tools (ovs-controller, ovs-ofctl, ovs-vsctl) that developers
92 can script and extend to provide distributed vswitch capabilities
93 that are closely integrated with their virtualization management
96 Q: Why doesn't Open vSwitch support distribution?
98 A: Open vSwitch is intended to be a useful component for building
99 flexible network infrastructure. There are many different approaches
100 to distribution which balance trade-offs between simplicity,
101 scalability, hardware compatibility, convergence times, logical
102 forwarding model, etc. The goal of Open vSwitch is to be able to
103 support all as a primitive building block rather than choose a
104 particular point in the distributed design space.
106 Q: How can I contribute to the Open vSwitch Community?
108 A: You can start by joining the mailing lists and helping to answer
109 questions. You can also suggest improvements to documentation. If
110 you have a feature or bug you would like to work on, send a mail to
111 one of the mailing lists:
113 http://openvswitch.org/mlists/
120 Q: What does it mean for an Open vSwitch release to be LTS (long-term
123 A: All official releases have been through a comprehensive testing
124 process and are suitable for production use. Planned releases will
125 occur several times a year. If a significant bug is identified in an
126 LTS release, we will provide an updated release that includes the
127 fix. Releases that are not LTS may not be fixed and may just be
128 supplanted by the next major release. The current LTS release is
131 Q: What Linux kernel versions does each Open vSwitch release work with?
133 A: The following table lists the Linux kernel versions against which the
134 given versions of the Open vSwitch kernel module will successfully
135 build. The Linux kernel versions are upstream kernel versions, so
136 Linux kernels modified from the upstream sources may not build in
137 some cases even if they are based on a supported version. This is
138 most notably true of Red Hat Enterprise Linux (RHEL) kernels, which
139 are extensively modified from upstream.
141 Open vSwitch Linux kernel
142 ------------ -------------
150 Open vSwitch userspace should also work with the Linux kernel module
151 built into Linux 3.3 and later.
153 Open vSwitch userspace is not sensitive to the Linux kernel version.
154 It should build against almost any kernel, certainly against 2.6.18
157 Q: What features are not available in the Open vSwitch kernel datapath
158 that ships as part of the upstream Linux kernel?
160 A: The kernel module in upstream Linux 3.3 and later does not include
161 the following features:
163 - Tunnel virtual ports, that is, interfaces with type "gre",
164 "ipsec_gre", "capwap". It is possible to create tunnels in
165 Linux and attach them to Open vSwitch as system devices.
166 However, they cannot be dynamically created through the OVSDB
167 protocol or set the tunnel ids as a flow action.
169 Work is in progress in adding these features to the upstream
170 Linux version of the Open vSwitch kernel module. For now, if
171 you need these features, use the kernel module from the Open
172 vSwitch distribution instead of the upstream Linux kernel
175 - Patch virtual ports, that is, interfaces with type "patch".
176 You can use Linux "veth" devices as a substitute.
178 We don't have any plans to add patch ports upstream.
180 Q: What features are not available when using the userspace datapath?
182 A: Tunnel and patch virtual ports are not supported, as described in the
183 previous answer. It is also not possible to use queue-related
184 actions. On Linux kernels before 2.6.39, maximum-sized VLAN packets
185 may not be transmitted.
191 Q: I thought Open vSwitch was a virtual Ethernet switch, but the
192 documentation keeps talking about bridges. What's a bridge?
194 A: In networking, the terms "bridge" and "switch" are synonyms. Open
195 vSwitch implements an Ethernet switch, which means that it is also
200 A: See the "VLAN" section below.
206 Q: How do I configure a port as an access port?
208 A: Add "tag=VLAN" to your "ovs-vsctl add-port" command. For example,
209 the following commands configure br0 with eth0 as a trunk port (the
210 default) and tap0 as an access port for VLAN 9:
213 ovs-vsctl add-port br0 eth0
214 ovs-vsctl add-port br0 tap0 tag=9
216 If you want to configure an already added port as an access port,
217 use "ovs-vsctl set", e.g.:
219 ovs-vsctl set port tap0 tag=9
221 Q: How do I configure a port as a SPAN port, that is, enable mirroring
222 of all traffic to that port?
224 A: The following commands configure br0 with eth0 and tap0 as trunk
225 ports. All traffic coming in or going out on eth0 or tap0 is also
226 mirrored to tap1; any traffic arriving on tap1 is dropped:
229 ovs-vsctl add-port br0 eth0
230 ovs-vsctl add-port br0 tap0
231 ovs-vsctl add-port br0 tap1 \
232 -- --id=@p get port tap1 \
233 -- --id=@m create mirror name=m0 select-all=true output-port=@p \
234 -- set bridge br0 mirrors=@m
236 To later disable mirroring, run:
238 ovs-vsctl clear bridge br0 mirrors
240 Q: How do I configure a VLAN as an RSPAN VLAN, that is, enable
241 mirroring of all traffic to that VLAN?
243 A: The following commands configure br0 with eth0 as a trunk port and
244 tap0 as an access port for VLAN 10. All traffic coming in or going
245 out on tap0, as well as traffic coming in or going out on eth0 in
246 VLAN 10, is also mirrored to VLAN 15 on eth0. The original tag for
247 VLAN 10, in cases where one is present, is dropped as part of
251 ovs-vsctl add-port br0 eth0
252 ovs-vsctl add-port br0 tap0 tag=10
254 -- --id=@m create mirror name=m0 select-all=true select-vlan=10 \
256 -- set bridge br0 mirrors=@m
258 To later disable mirroring, run:
260 ovs-vsctl clear bridge br0 mirrors
262 Mirroring to a VLAN can disrupt a network that contains unmanaged
263 switches. See ovs-vswitchd.conf.db(5) for details. Mirroring to a
264 GRE tunnel has fewer caveats than mirroring to a VLAN and should
265 generally be preferred.
267 Q: Can I mirror more than one input VLAN to an RSPAN VLAN?
269 A: Yes, but mirroring to a VLAN strips the original VLAN tag in favor
270 of the specified output-vlan. This loss of information may make
271 the mirrored traffic too hard to interpret.
273 To mirror multiple VLANs, use the commands above, but specify a
274 comma-separated list of VLANs as the value for select-vlan. To
275 mirror every VLAN, use the commands above, but omit select-vlan and
278 When a packet arrives on a VLAN that is used as a mirror output
279 VLAN, the mirror is disregarded. Instead, in standalone mode, OVS
280 floods the packet across all the ports for which the mirror output
281 VLAN is configured. (If an OpenFlow controller is in use, then it
282 can override this behavior through the flow table.) If OVS is used
283 as an intermediate switch, rather than an edge switch, this ensures
284 that the RSPAN traffic is distributed through the network.
286 Mirroring to a VLAN can disrupt a network that contains unmanaged
287 switches. See ovs-vswitchd.conf.db(5) for details. Mirroring to a
288 GRE tunnel has fewer caveats than mirroring to a VLAN and should
289 generally be preferred.
291 Q: How do I configure mirroring of all traffic to a GRE tunnel?
293 A: The following commands configure br0 with eth0 and tap0 as trunk
294 ports. All traffic coming in or going out on eth0 or tap0 is also
295 mirrored to gre0, a GRE tunnel to the remote host 192.168.1.10; any
296 traffic arriving on gre0 is dropped:
299 ovs-vsctl add-port br0 eth0
300 ovs-vsctl add-port br0 tap0
301 ovs-vsctl add-port br0 gre0 \
302 -- set interface gre0 type=gre options:remote_ip=192.168.1.10 \
303 -- --id=@p get port gre0 \
304 -- --id=@m create mirror name=m0 select-all=true output-port=@p \
305 -- set bridge br0 mirrors=@m
307 To later disable mirroring and destroy the GRE tunnel:
309 ovs-vsctl clear bridge br0 mirrors
310 ovs-vcstl del-port br0 gre0
312 Q: Does Open vSwitch support ERSPAN?
314 A: No. ERSPAN is an undocumented proprietary protocol. As an
315 alternative, Open vSwitch supports mirroring to a GRE tunnel (see
319 Configuration Problems
320 ----------------------
322 Q: I created a bridge and added my Ethernet port to it, using commands
326 ovs-vsctl add-port br0 eth0
328 and as soon as I ran the "add-port" command I lost all connectivity
331 A: A physical Ethernet device that is part of an Open vSwitch bridge
332 should not have an IP address. If one does, then that IP address
333 will not be fully functional.
335 You can restore functionality by moving the IP address to an Open
336 vSwitch "internal" device, such as the network device named after
337 the bridge itself. For example, assuming that eth0's IP address is
338 192.168.128.5, you could run the commands below to fix up the
341 ifconfig eth0 0.0.0.0
342 ifconfig br0 192.168.128.5
344 (If your only connection to the machine running OVS is through the
345 IP address in question, then you would want to run all of these
346 commands on a single command line, or put them into a script.) If
347 there were any additional routes assigned to eth0, then you would
348 also want to use commands to adjust these routes to go through br0.
350 If you use DHCP to obtain an IP address, then you should kill the
351 DHCP client that was listening on the physical Ethernet interface
352 (e.g. eth0) and start one listening on the internal interface
353 (e.g. br0). You might still need to manually clear the IP address
354 from the physical interface (e.g. with "ifconfig eth0 0.0.0.0").
356 There is no compelling reason why Open vSwitch must work this way.
357 However, this is the way that the Linux kernel bridge module has
358 always worked, so it's a model that those accustomed to Linux
359 bridging are already used to. Also, the model that most people
360 expect is not implementable without kernel changes on all the
361 versions of Linux that Open vSwitch supports.
363 By the way, this issue is not specific to physical Ethernet
364 devices. It applies to all network devices except Open vswitch
367 Q: I created a bridge and added a couple of Ethernet ports to it,
368 using commands like these:
371 ovs-vsctl add-port br0 eth0
372 ovs-vsctl add-port br0 eth1
374 and now my network seems to have melted: connectivity is unreliable
375 (even connectivity that doesn't go through Open vSwitch), all the
376 LEDs on my physical switches are blinking, wireshark shows
377 duplicated packets, and CPU usage is very high.
379 A: More than likely, you've looped your network. Probably, eth0 and
380 eth1 are connected to the same physical Ethernet switch. This
381 yields a scenario where OVS receives a broadcast packet on eth0 and
382 sends it out on eth1, then the physical switch connected to eth1
383 sends the packet back on eth0, and so on forever. More complicated
384 scenarios, involving a loop through multiple switches, are possible
387 The solution depends on what you are trying to do:
389 - If you added eth0 and eth1 to get higher bandwidth or higher
390 reliability between OVS and your physical Ethernet switch,
391 use a bond. The following commands create br0 and then add
392 eth0 and eth1 as a bond:
395 ovs-vsctl add-bond br0 bond0 eth0 eth1
397 Bonds have tons of configuration options. Please read the
398 documentation on the Port table in ovs-vswitchd.conf.db(5)
401 - Perhaps you don't actually need eth0 and eth1 to be on the
402 same bridge. For example, if you simply want to be able to
403 connect each of them to virtual machines, then you can put
404 each of them on a bridge of its own:
407 ovs-vsctl add-port br0 eth0
410 ovs-vsctl add-port br1 eth1
412 and then connect VMs to br0 and br1. (A potential
413 disadvantage is that traffic cannot directly pass between br0
414 and br1. Instead, it will go out eth0 and come back in eth1,
417 - If you have a redundant or complex network topology and you
418 want to prevent loops, turn on spanning tree protocol (STP).
419 The following commands create br0, enable STP, and add eth0
420 and eth1 to the bridge. The order is important because you
421 don't want have to have a loop in your network even
425 ovs-vsctl set bridge br0 stp_enable=true
426 ovs-vsctl add-port br0 eth0
427 ovs-vsctl add-port br0 eth1
429 The Open vSwitch implementation of STP is not well tested.
430 Please report any bugs you observe, but if you'd rather avoid
431 acting as a beta tester then another option might be your
434 Q: I can't seem to use Open vSwitch in a wireless network.
436 A: Wireless base stations generally only allow packets with the source
437 MAC address of NIC that completed the initial handshake.
438 Therefore, without MAC rewriting, only a single device can
439 communicate over a single wireless link.
441 This isn't specific to Open vSwitch, it's enforced by the access
442 point, so the same problems will show up with the Linux bridge or
443 any other way to do bridging.
445 Q: Is there any documentation on the database tables and fields?
447 A: Yes. ovs-vswitchd.conf.db(5) is a comprehensive reference.
449 Q: When I run ovs-dpctl I no longer see the bridges I created. Instead,
450 I only see a datapath called "ovs-system". How can I see datapath
451 information about a particular bridge?
453 A: In version 1.9.0, OVS switched to using a single datapath that is
454 shared by all bridges of that type. The "ovs-appctl dpif/*"
455 commands provide similar functionality that is scoped by the bridge.
458 Quality of Service (QoS)
459 ------------------------
461 Q: How do I configure Quality of Service (QoS)?
463 A: Suppose that you want to set up bridge br0 connected to physical
464 Ethernet port eth0 (a 1 Gbps device) and virtual machine interfaces
465 vif1.0 and vif2.0, and that you want to limit traffic from vif1.0
466 to eth0 to 10 Mbps and from vif2.0 to eth0 to 20 Mbps. Then, you
467 could configure the bridge this way:
471 add-port br0 eth0 -- \
472 add-port br0 vif1.0 -- set interface vif1.0 ofport_request=5 -- \
473 add-port br0 vif2.0 -- set interface vif2.0 ofport_request=6 -- \
474 set port eth0 qos=@newqos -- \
475 --id=@newqos create qos type=linux-htb \
476 other-config:max-rate=1000000000 \
477 queues:123=@vif10queue \
478 queues:234=@vif20queue -- \
479 --id=@vif10queue create queue other-config:max-rate=10000000 -- \
480 --id=@vif20queue create queue other-config:max-rate=20000000
482 At this point, bridge br0 is configured with the ports and eth0 is
483 configured with the queues that you need for QoS, but nothing is
484 actually directing packets from vif1.0 or vif2.0 to the queues that
485 we have set up for them. That means that all of the packets to
486 eth0 are going to the "default queue", which is not what we want.
488 We use OpenFlow to direct packets from vif1.0 and vif2.0 to the
489 queues reserved for them:
491 ovs-ofctl add-flow br0 in_port=5,actions=set_queue:123,normal
492 ovs-ofctl add-flow br0 in_port=6,actions=set_queue:234,normal
494 Each of the above flows matches on the input port, sets up the
495 appropriate queue (123 for vif1.0, 234 for vif2.0), and then
496 executes the "normal" action, which performs the same switching
497 that Open vSwitch would have done without any OpenFlow flows being
498 present. (We know that vif1.0 and vif2.0 have OpenFlow port
499 numbers 5 and 6, respectively, because we set their ofport_request
500 columns above. If we had not done that, then we would have needed
501 to find out their port numbers before setting up these flows.)
503 Now traffic going from vif1.0 or vif2.0 to eth0 should be
506 By the way, if you delete the bridge created by the above commands,
511 then that will leave one unreferenced QoS record and two
512 unreferenced Queue records in the Open vSwich database. One way to
513 clear them out, assuming you don't have other QoS or Queue records
514 that you want to keep, is:
516 ovs-vsctl -- --all destroy QoS -- --all destroy Queue
518 Q: I configured Quality of Service (QoS) in my OpenFlow network by
519 adding records to the QoS and Queue table, but the results aren't
522 A: Did you install OpenFlow flows that use your queues? This is the
523 primary way to tell Open vSwitch which queues you want to use. If
524 you don't do this, then the default queue will be used, which will
525 probably not have the effect you want.
527 Refer to the previous question for an example.
529 Q: I configured QoS, correctly, but my measurements show that it isn't
530 working as well as I expect.
532 A: With the Linux kernel, the Open vSwitch implementation of QoS has
535 - Open vSwitch configures a subset of Linux kernel QoS
536 features, according to what is in OVSDB. It is possible that
537 this code has bugs. If you believe that this is so, then you
538 can configure the Linux traffic control (QoS) stack directly
539 with the "tc" program. If you get better results that way,
540 you can send a detailed bug report to bugs@openvswitch.org.
542 It is certain that Open vSwitch cannot configure every Linux
543 kernel QoS feature. If you need some feature that OVS cannot
544 configure, then you can also use "tc" directly (or add that
547 - The Open vSwitch implementation of OpenFlow allows flows to
548 be directed to particular queues. This is pretty simple and
549 unlikely to have serious bugs at this point.
551 However, most problems with QoS on Linux are not bugs in Open
552 vSwitch at all. They tend to be either configuration errors
553 (please see the earlier questions in this section) or issues with
554 the traffic control (QoS) stack in Linux. The Open vSwitch
555 developers are not experts on Linux traffic control. We suggest
556 that, if you believe you are encountering a problem with Linux
557 traffic control, that you consult the tc manpages (e.g. tc(8),
558 tc-htb(8), tc-hfsc(8)), web resources (e.g. http://lartc.org/), or
559 mailing lists (e.g. http://vger.kernel.org/vger-lists.html#netdev).
567 A: At the simplest level, a VLAN (short for "virtual LAN") is a way to
568 partition a single switch into multiple switches. Suppose, for
569 example, that you have two groups of machines, group A and group B.
570 You want the machines in group A to be able to talk to each other,
571 and you want the machine in group B to be able to talk to each
572 other, but you don't want the machines in group A to be able to
573 talk to the machines in group B. You can do this with two
574 switches, by plugging the machines in group A into one switch and
575 the machines in group B into the other switch.
577 If you only have one switch, then you can use VLANs to do the same
578 thing, by configuring the ports for machines in group A as VLAN
579 "access ports" for one VLAN and the ports for group B as "access
580 ports" for a different VLAN. The switch will only forward packets
581 between ports that are assigned to the same VLAN, so this
582 effectively subdivides your single switch into two independent
583 switches, one for each group of machines.
585 So far we haven't said anything about VLAN headers. With access
586 ports, like we've described so far, no VLAN header is present in
587 the Ethernet frame. This means that the machines (or switches)
588 connected to access ports need not be aware that VLANs are
589 involved, just like in the case where we use two different physical
592 Now suppose that you have a whole bunch of switches in your
593 network, instead of just one, and that some machines in group A are
594 connected directly to both switches 1 and 2. To allow these
595 machines to talk to each other, you could add an access port for
596 group A's VLAN to switch 1 and another to switch 2, and then
597 connect an Ethernet cable between those ports. That works fine,
598 but it doesn't scale well as the number of switches and the number
599 of VLANs increases, because you use up a lot of valuable switch
600 ports just connecting together your VLANs.
602 This is where VLAN headers come in. Instead of using one cable and
603 two ports per VLAN to connect a pair of switches, we configure a
604 port on each switch as a VLAN "trunk port". Packets sent and
605 received on a trunk port carry a VLAN header that says what VLAN
606 the packet belongs to, so that only two ports total are required to
607 connect the switches, regardless of the number of VLANs in use.
608 Normally, only switches (either physical or virtual) are connected
609 to a trunk port, not individual hosts, because individual hosts
610 don't expect to see a VLAN header in the traffic that they receive.
612 None of the above discussion says anything about particular VLAN
613 numbers. This is because VLAN numbers are completely arbitrary.
614 One must only ensure that a given VLAN is numbered consistently
615 throughout a network and that different VLANs are given different
616 numbers. (That said, VLAN 0 is usually synonymous with a packet
617 that has no VLAN header, and VLAN 4095 is reserved.)
621 A: Many drivers in Linux kernels before version 3.3 had VLAN-related
622 bugs. If you are having problems with VLANs that you suspect to be
623 driver related, then you have several options:
625 - Upgrade to Linux 3.3 or later.
627 - Build and install a fixed version of the particular driver
628 that is causing trouble, if one is available.
630 - Use a NIC whose driver does not have VLAN problems.
632 - Use "VLAN splinters", a feature in Open vSwitch 1.4 and later
633 that works around bugs in kernel drivers. To enable VLAN
634 splinters on interface eth0, use the command:
636 ovs-vsctl set interface eth0 other-config:enable-vlan-splinters=true
638 For VLAN splinters to be effective, Open vSwitch must know
639 which VLANs are in use. See the "VLAN splinters" section in
640 the Interface table in ovs-vswitchd.conf.db(5) for details on
641 how Open vSwitch infers in-use VLANs.
643 VLAN splinters increase memory use and reduce performance, so
644 use them only if needed.
646 - Apply the "vlan workaround" patch from the XenServer kernel
647 patch queue, build Open vSwitch against this patched kernel,
648 and then use ovs-vlan-bug-workaround(8) to enable the VLAN
649 workaround for each interface whose driver is buggy.
651 (This is a nontrivial exercise, so this option is included
652 only for completeness.)
654 It is not always easy to tell whether a Linux kernel driver has
655 buggy VLAN support. The ovs-vlan-test(8) and ovs-test(8) utilities
656 can help you test. See their manpages for details. Of the two
657 utilities, ovs-test(8) is newer and more thorough, but
658 ovs-vlan-test(8) may be easier to use.
660 Q: VLANs still don't work. I've tested the driver so I know that it's OK.
662 A: Do you have VLANs enabled on the physical switch that OVS is
663 attached to? Make sure that the port is configured to trunk the
664 VLAN or VLANs that you are using with OVS.
666 Q: Outgoing VLAN-tagged traffic goes through OVS to my physical switch
667 and to its destination host, but OVS seems to drop incoming return
670 A: It's possible that you have the VLAN configured on your physical
671 switch as the "native" VLAN. In this mode, the switch treats
672 incoming packets either tagged with the native VLAN or untagged as
673 part of the native VLAN. It may also send outgoing packets in the
674 native VLAN without a VLAN tag.
676 If this is the case, you have two choices:
678 - Change the physical switch port configuration to tag packets
679 it forwards to OVS with the native VLAN instead of forwarding
682 - Change the OVS configuration for the physical port to a
683 native VLAN mode. For example, the following sets up a
684 bridge with port eth0 in "native-tagged" mode in VLAN 9:
687 ovs-vsctl add-port br0 eth0 tag=9 vlan_mode=native-tagged
689 In this situation, "native-untagged" mode will probably work
690 equally well. Refer to the documentation for the Port table
691 in ovs-vswitchd.conf.db(5) for more information.
693 Q: I added a pair of VMs on different VLANs, like this:
696 ovs-vsctl add-port br0 eth0
697 ovs-vsctl add-port br0 tap0 tag=9
698 ovs-vsctl add-port br0 tap1 tag=10
700 but the VMs can't access each other, the external network, or the
703 A: It is to be expected that the VMs can't access each other. VLANs
704 are a means to partition a network. When you configured tap0 and
705 tap1 as access ports for different VLANs, you indicated that they
706 should be isolated from each other.
708 As for the external network and the Internet, it seems likely that
709 the machines you are trying to access are not on VLAN 9 (or 10) and
710 that the Internet is not available on VLAN 9 (or 10).
712 Q: Can I configure an IP address on a VLAN?
714 A: Yes. Use an "internal port" configured as an access port. For
715 example, the following configures IP address 192.168.0.7 on VLAN 9.
716 That is, OVS will forward packets from eth0 to 192.168.0.7 only if
717 they have an 802.1Q header with VLAN 9. Conversely, traffic
718 forwarded from 192.168.0.7 to eth0 will be tagged with an 802.1Q
722 ovs-vsctl add-port br0 eth0
723 ovs-vsctl add-port br0 vlan9 tag=9 -- set interface vlan9 type=internal
724 ifconfig vlan9 192.168.0.7
726 Q: My OpenFlow controller doesn't see the VLANs that I expect.
728 A: The configuration for VLANs in the Open vSwitch database (e.g. via
729 ovs-vsctl) only affects traffic that goes through Open vSwitch's
730 implementation of the OpenFlow "normal switching" action. By
731 default, when Open vSwitch isn't connected to a controller and
732 nothing has been manually configured in the flow table, all traffic
733 goes through the "normal switching" action. But, if you set up
734 OpenFlow flows on your own, through a controller or using ovs-ofctl
735 or through other means, then you have to implement VLAN handling
738 You can use "normal switching" as a component of your OpenFlow
739 actions, e.g. by putting "normal" into the lists of actions on
740 ovs-ofctl or by outputting to OFPP_NORMAL from an OpenFlow
741 controller. This will only be suitable for some situations,
744 Q: I configured ports on a bridge as access ports with different VLAN
748 ovs-vsctl set-controller br0 tcp:192.168.0.10:6633
749 ovs-vsctl add-port br0 eth0
750 ovs-vsctl add-port br0 tap0 tag=9
751 ovs-vsctl add-port br0 tap1 tag=10
753 but the VMs running behind tap0 and tap1 can still communicate,
754 that is, they are not isolated from each other even though they are
757 A: Do you have a controller configured on br0 (as the commands above
758 do)? If so, then this is a variant on the previous question, "My
759 OpenFlow controller doesn't see the VLANs that I expect," and you
760 can refer to the answer there for more information.
766 Q: What versions of OpenFlow does Open vSwitch support?
768 A: Open vSwitch 1.9 and earlier support only OpenFlow 1.0 (plus
769 extensions that bring in many of the features from later versions
772 Open vSwitch versions 1.10 and later will have experimental support
773 for OpenFlow 1.2 and 1.3. On these versions of Open vSwitch, the
774 following command enables OpenFlow 1.0, 1.2, and 1.3 on bridge br0:
776 ovs-vsctl set bridge br0 protocols=openflow10,openflow12,openflow13
778 Support for OpenFlow 1.1 is incomplete enough that it cannot yet be
779 enabled, even experimentally.
781 Support for OpenFlow 1.2 and 1.3 is still incomplete. Work to be
782 done is tracked in OPENFLOW-1.1+ in the Open vSwitch source tree
783 (also via http://openvswitch.org/development/openflow-1-x-plan/).
784 When support for a given OpenFlow version is solidly implemented,
785 Open vSwitch will enable that version by default.
787 Q: I'm getting "error type 45250 code 0". What's that?
789 A: This is a Open vSwitch extension to OpenFlow error codes. Open
790 vSwitch uses this extension when it must report an error to an
791 OpenFlow controller but no standard OpenFlow error code is
794 Open vSwitch logs the errors that it sends to controllers, so the
795 easiest thing to do is probably to look at the ovs-vswitchd log to
796 find out what the error was.
798 If you want to dissect the extended error message yourself, the
799 format is documented in include/openflow/nicira-ext.h in the Open
800 vSwitch source distribution. The extended error codes are
801 documented in lib/ofp-errors.h.
803 Q1: Some of the traffic that I'd expect my OpenFlow controller to see
804 doesn't actually appear through the OpenFlow connection, even
805 though I know that it's going through.
806 Q2: Some of the OpenFlow flows that my controller sets up don't seem
807 to apply to certain traffic, especially traffic between OVS and
808 the controller itself.
810 A: By default, Open vSwitch assumes that OpenFlow controllers are
811 connected "in-band", that is, that the controllers are actually
812 part of the network that is being controlled. In in-band mode,
813 Open vSwitch sets up special "hidden" flows to make sure that
814 traffic can make it back and forth between OVS and the controllers.
815 These hidden flows are higher priority than any flows that can be
816 set up through OpenFlow, and they are not visible through normal
817 OpenFlow flow table dumps.
819 Usually, the hidden flows are desirable and helpful, but
820 occasionally they can cause unexpected behavior. You can view the
821 full OpenFlow flow table, including hidden flows, on bridge br0
824 ovs-appctl bridge/dump-flows br0
826 to help you debug. The hidden flows are those with priorities
827 greater than 65535 (the maximum priority that can be set with
830 The DESIGN file at the top level of the Open vSwitch source
831 distribution describes the in-band model in detail.
833 If your controllers are not actually in-band (e.g. they are on
834 localhost via 127.0.0.1, or on a separate network), then you should
835 configure your controllers in "out-of-band" mode. If you have one
836 controller on bridge br0, then you can configure out-of-band mode
839 ovs-vsctl set controller br0 connection-mode=out-of-band
841 Q: I configured all my controllers for out-of-band control mode but
842 "ovs-appctl bridge/dump-flows" still shows some hidden flows.
844 A: You probably have a remote manager configured (e.g. with "ovs-vsctl
845 set-manager"). By default, Open vSwitch assumes that managers need
846 in-band rules set up on every bridge. You can disable these rules
849 ovs-vsctl set bridge br0 other-config:disable-in-band=true
851 This actually disables in-band control entirely for the bridge, as
852 if all the bridge's controllers were configured for out-of-band
855 Q: My OpenFlow controller doesn't see the VLANs that I expect.
857 A: See answer under "VLANs", above.
859 Q: I ran "ovs-ofctl add-flow br0 nw_dst=192.168.0.1,actions=drop"
860 but I got a funny message like this:
862 ofp_util|INFO|normalization changed ofp_match, details:
863 ofp_util|INFO| pre: nw_dst=192.168.0.1
866 and when I ran "ovs-ofctl dump-flows br0" I saw that my nw_dst
867 match had disappeared, so that the flow ends up matching every
870 A: The term "normalization" in the log message means that a flow
871 cannot match on an L3 field without saying what L3 protocol is in
872 use. The "ovs-ofctl" command above didn't specify an L3 protocol,
873 so the L3 field match was dropped.
875 In this case, the L3 protocol could be IP or ARP. A correct
876 command for each possibility is, respectively:
878 ovs-ofctl add-flow br0 ip,nw_dst=192.168.0.1,actions=drop
882 ovs-ofctl add-flow br0 arp,nw_dst=192.168.0.1,actions=drop
884 Similarly, a flow cannot match on an L4 field without saying what
885 L4 protocol is in use. For example, the flow match "tp_src=1234"
886 is, by itself, meaningless and will be ignored. Instead, to match
887 TCP source port 1234, write "tcp,tp_src=1234", or to match UDP
888 source port 1234, write "udp,tp_src=1234".
890 Q: How can I figure out the OpenFlow port number for a given port?
892 A: The OFPT_FEATURES_REQUEST message requests an OpenFlow switch to
893 respond with an OFPT_FEATURES_REPLY that, among other information,
894 includes a mapping between OpenFlow port names and numbers. From a
895 command prompt, "ovs-ofctl show br0" makes such a request and
896 prints the response for switch br0.
898 The Interface table in the Open vSwitch database also maps OpenFlow
899 port names to numbers. To print the OpenFlow port number
900 associated with interface eth0, run:
902 ovs-vsctl get Interface eth0 ofport
904 You can print the entire mapping with:
906 ovs-vsctl -- --columns=name,ofport list Interface
908 but the output mixes together interfaces from all bridges in the
909 database, so it may be confusing if more than one bridge exists.
911 In the Open vSwitch database, ofport value -1 means that the
912 interface could not be created due to an error. (The Open vSwitch
913 log should indicate the reason.) ofport value [] (the empty set)
914 means that the interface hasn't been created yet. The latter is
915 normally an intermittent condition (unless ovs-vswitchd is not
918 Q: I added some flows with my controller or with ovs-ofctl, but when I
919 run "ovs-dpctl dump-flows" I don't see them.
921 A: ovs-dpctl queries a kernel datapath, not an OpenFlow switch. It
922 won't display the information that you want. You want to use
923 "ovs-ofctl dump-flows" instead.
925 Q: It looks like each of the interfaces in my bonded port shows up
926 as an individual OpenFlow port. Is that right?
928 A: Yes, Open vSwitch makes individual bond interfaces visible as
929 OpenFlow ports, rather than the bond as a whole. The interfaces
930 are treated together as a bond for only a few purposes:
932 - Sending a packet to the OFPP_NORMAL port. (When an OpenFlow
933 controller is not configured, this happens implicitly to
936 - The "autopath" Nicira extension action. However, "autopath"
937 is deprecated and scheduled for removal in February 2013.
939 - Mirrors configured for output to a bonded port.
941 It would make a lot of sense for Open vSwitch to present a bond as
942 a single OpenFlow port. If you want to contribute an
943 implementation of such a feature, please bring it up on the Open
944 vSwitch development mailing list at dev@openvswitch.org.
950 http://openvswitch.org/
git://git.onelab.eu / sliver-openvswitch.git / blob